New type of antibiotic discovered by artificial intelligence could defeat dangerous superbugs

CNN
—

Researchers say they have used artificial intelligence to discover a new class of antibiotics that are particularly effective against threatening drug-resistant bacteria.

When the antibiotic was tested on the skin of mice ly infected with superbugs, it inhibited bacterial growth, suggesting that the method could be used to create tailored antibiotics to combat other drug-resistant pathogens. suggested that it could be used.

Additionally, the compounds identified by the AI ​​functioned to stop only the pathogen in question. It does not appear to kill many other species of beneficial bacteria that live in the gut or on the skin, making it a rare narrowly targeted drug.

The researchers said that more antibiotics working so precisely could prevent bacteria from becoming resistant in the first place.

the study Published It was published in Nature Chemical Biology.

“This is very promising,” said Dr. Cesar de la Fuente, an assistant professor at the University of Pennsylvania’s Perlman School of Medicine. Although he is using AI to find new treatments, he is not involved in any new research.

De la Fuente said this kind of approach to discovering new drugs is an emerging field that researchers have been experimenting with since around 2018. This greatly reduces the time it takes to screen thousands of promising compounds.

“As we have seen, I think AI can be applied well in many areas. I think drug discovery is kind of the next frontier.”

In this study, the researchers focused on the bacterium Actinetobacter baumannii. It inhabits hospitals and other medical settings, clinging to surfaces such as doorknobs and counters. Being able to grab snippets of DNA from other organisms they come in contact with, they can incorporate genes that help them resist the drugs doctors use to treat them, their best weapon.

“This is what we call professional pathogens in the lab,” said John Stokes, one of the researchers and an assistant professor of biochemistry and biomedical sciences at McMaster University in Hamilton, Ontario.

This species causes skin, blood and respiratory infections that are difficult to treat.U.S. Centers for Disease Control and Prevention said in 2019 That Acinetobacter baumannii New types of antibiotics are “most needed” to treat infections.

a Recent research One in four hospitalized patients with Actinetobacter baumannii infections that were also resistant to strong carbapenem antibiotics were found to have died within one month of diagnosis.

For the new research, Dr. Stokes and his lab collaborated with researchers at MIT and the Broad Institute at Harvard University. First, they grew Acinetobacter baumannii in laboratory dishes using a technique called high-throughput drug screening and spent weeks exposing these colonies to more than 7,500 drugs (drugs and drug active ingredients). spent. They discovered 480 compounds that inhibit bacterial growth.

They put that information into a computer and used it to train an artificial intelligence algorithm.

“Once you have trained a model, the next thing you can do is start showing it a whole new picture of chemicals that it has never seen before, right? It predicts whether something is antibacterial,” Stokes said.

They then had the model screen more than 6,000 molecules, which Stokes said the AI ​​was able to do over hours.

They narrowed their search to 240 chemicals and tested them in the lab. Laboratory tests allowed us to narrow down the list of the best inhibitors against bacteria to nine. From there, they took a closer look at each structure and ruled out those they thought could be dangerous or related to known antibiotics.

They were left with one compound called RS102895, which Stokes believes was originally developed as a potential treatment for diabetes.

He said it appears to work in a completely new way, preventing components of the bacterium from moving from inside the cell to the cell surface.

“It’s a pretty interesting mechanism, one that, to my knowledge, has never been observed with clinical antibiotics,” he said.

In addition, RS102895, which the researchers renamed Abaucin, works only on Actinetobacter baumannii, he said.

Stokes said most antibiotics are broad-spectrum drugs and work against many types of bacteria. Broad-spectrum antibiotics put tremendous selective pressure on many types of bacteria, causing them to rapidly evolve and share genes that help many bacteria resist and survive drugs.

“For this molecule, we can’t impose a universal selective pressure because it’s only acting so strongly against Actinetobacter, so resistance doesn’t spread that quickly,” he said.